The identification of separate disease 'subtypes' or 'intermediate phenotypes' within the syndrome of schizophrenia would facilitate future research on etiology, identification of salient genes and biological markers, and enable more effective prevention and intervention. One promising schizophrenia 'subtype', the 'deficit syndrome' (DS), is characterized by persistent and primary negative symptoms. The deficit syndrome differs from general negative symptoms in its emphasis on primary negative symptoms, predictive power for outcomes, and specific risk factors. While a substantial literature supports the differentiation of 'deficit' from 'non-deficit' syndrome schizophrenia, there is an important gap in the evidence, which this proposal seeks to address. Methodological concerns and geographic limitations of studies have not allowed researchers to fully answer three important questions concerning this subtype: 1) effects of acute psychosis; 2) potential effects of medication treatment; 3) cross-cultural generalizability. In particular, no studies to date have effectively ruled out both the effects of acute psychosis and prolonged medication treatment in the assessment of the deficit syndrome. This proposal seeks to address this gap via a secondary data analysis of the sole existing representative, non-acute, population-based, sample of 'untreated or minimally-treated' schizophrenia patients obtained from a landmark psychiatric epidemiology study in a non-Western context (China). Utilizing this population-based sample of chronic, untreated psychotic illness--who have had untreated illness for an average of 10.5 years and are thus likely to have distinct clinical features such as greater symptomatology--offers an extraordinary test for construct validity of the deficit syndrome. We propose to utilize a representative sample of 389 patients diagnosed with psychotic disorders obtained from a stratified random sample of 4 provinces in China (a sampling frame of 113 million adults). This sample offers unique advantages over prior studies in its: 1) large size (n=389); 2) representative, population-based sampling with data obtained via a clinician-administered interview; and 3) large numbers of 'untreated/minimally-treated' patients (n= 208), thus affording a unique opportunity to comprehensively test the validity of the 'deficit syndrome' subtype within a non-acute, representatively-sampled 'untreated and minimally-treated group' in China. We first seek to confirm the deficit syndrome construct in a group of psychotic patients 'substantially exposed to medication treatment' in this sample (i.e., havinge1 psychiatric hospitalizations), then to assess the construct validity of deficit syndrome within a unique 'untreated or minimally treated' group of psychotic patients. Our specific aims are: 1) To identify cases of deficit syndrome and assess construct validity of deficit syndrome among 'treated' patients with psychotic disorders in China to determine whether this subtype demonstrates the same pattern of correlations to key demographic and clinical variables already established in Western samples. We control for developmental effects of illness in all analyses. Confirming the construct validity of deficit syndrome among the treated group in China will establish a baseline condition to test Aim #2. 2) To identify cases of DS and assess construct validity of deficit syndrome among the 'untreated or minimally treated' psychotic patient group to determine whether this subtype shows the same pattern of correlations to key demographic and clinical variables as the 'treated' group in China. 3) To explore among the full sample (n= 389) whether the relationship between the two treatment groups and key construct validation variables differ in magnitude by group. To model statistical interaction of 'deficit syndrome categorization' X 'treatment group status', we will examine whether the distribution of our variables used to demonstrate construct validity between DS and non-DS varies by treatment groups This will be the first study to utilize an untreated, non-acute, population-based sample of schizophrenia patients to establish construct validity for deficit syndrome, thereby controlling for acute psychosis, exposure to medication use, and validating deficit syndrome in a representative, non-Western based sample. This study has promise to provide evidence for deficit syndrome as a distinct disease 'subtype' within schizophrenia, thus contributing to future etiological, genetic linkage, and intervention studies in schizophrenia. This R03 (small grant) will be the first of several proposals from the Global Mental Health Program at Columbia that will initiate a novel and productive program of research examining subtypes and course determinants of untreated schizophrenia in non-Western societies to examine schizophrenia's cross-cultural aspects.